This Week’s Awesome Tech Stories From Around the Web (Through October 5)

MIT Spinoff Liquid Debuts Non-Transformer AI Models and They’re Already State-of-the-Art
Carl Franzen | VentureBeat
“Unlike most others of the current generative AI wave, these models are not based around the transformer architecture outlined in the seminal 2017 paper ‘Attention Is All You Need.’ Instead, Liquid states that its goal ‘is to explore ways to build foundation models beyond Generative Pre-trained Transformers (GPTs)’ and with the new LFMs, specifically building from ‘first principles…the same way engineers built engines, cars, and airplanes.’”

In Medical First, Woman’s Type 1 Diabetes Seemingly Cured by Stem Cells
Ed Cara | Gizmodo
“We might someday be able to have replacement insulin-making cells on demand. Scientists in China have presented early clinical trial data suggesting that a person’s stem cells can be turned into a steady supply of the pancreatic cells responsible for producing insulin. If truly successful, such a treatment would essentially cure Type 1 diabetes.”

AI-Generated Images Can Teach Robots How to Act
Rhiannon Williams | MIT Technology Review
“Researchers from Stephen James’s Robot Learning Lab in London are using image-generating AI models for a new purpose: creating training data for robots. They’ve developed a new system, called Genima, that fine-tunes the image-generating AI model Stable Diffusion to draw robots’ movements, helping guide them both in simulations and in the real world.”

Britain Shuts Down Last Coal Plant, ‘Turning Its Back on Coal Forever’
Somini Sengupta | The New York Times
“Britain, the nation that launched a global addiction to coal 150 years ago, is shutting down its last coal-burning power station on Monday. That makes Britain first among the world’s major, industrialized economies to wean itself off coal—all the more symbolic because it was also the first to burn tremendous amounts of it to fuel the Industrial Revolution, inspiring the rest of the world to follow suit.”

Four-Legged Robot Learns to Climb Ladders
Brian Heater | TechCrunch
“The proliferation of robots like Boston Dynamics’ Spot has showcased the versatility of quadrupeds. These systems have thrived at walking up stairs, traversing small obstacles, and navigating uneven terrain. Ladders, however, still present a big issue — especially given how ever present they are in factories and other industrial environments where the systems are deployed.”

Eight Scientists, a Billion Dollars, and the Moonshot Agency Trying to Make Britain Great Again
Matt Reynolds | Wired
“The whole point of ARIA is to push researchers beyond their comfort zones and towards ideas the typically risk-averse British science funding system would deem improbable or downright weird. …The plan should be just on the edge of impossible, Gur tells the room. Impactful enough that it’s worth a shot, but so ambitious that half of the scientists leave the workshop convinced it’ll never work.”

An ‘iPhone of AI’ Makes No Sense. What Is Jony Ive Really Building?
Sophie Charara | Wired
“LoveFrom is working with OpenAI to build AI devices that are less ‘socially disruptive’ than the iPhone. Is Ive looking for absolution or a new computing soul? …There is the sense…that LoveFrom has Apple-level talent, as close as it will get to Apple-level money—with plans to raise as much as $1 billion in funding by the end of this year—and, with Sam Altman involved, Apple-level ambitions.”

ART

Hidden ‘BopSpotter’ Microphone Is Constantly Surveilling San Francisco for Good Music
Jason Koebler | 404 Media
“Bop Spotter is a project by technologist Riley Walz in which he has hidden an Android phone in a box on a pole, rigged it to be solar powered, and has set it to record audio and periodically sends it to Shazam’s API to determine which songs people are playing in public. Walz describes it as ShotSpotter, but for music. ‘This is culture surveillance. No one notices, no one consents. But it’s not about catching criminals,’ Walz’s website reads. ‘It’s about catching vibes.’”

License Plate Readers Are Creating a US-Wide Database of More Than Just Cars
Matt Burgess and Dhruv Mehrotra | Wired
“Beyond highlighting the far-reaching nature of LPR technology, which has collected billions of images of license plates, the research also shows how people’s personal political views and their homes can be recorded into vast databases that can be queried. ‘It really reveals the extent to which surveillance is happening on a mass scale in the quiet streets of America,’ says Jay Stanley, a senior policy analyst at the American Civil Liberties Union.”

Moon Time Is a Thing Now—Here’s Why It Matters
Rebecca Boyle | Atlas Obscura
“Moon time is a meaningful thing to understand, especially as countries and private companies are angling to return to the lunar surface this decade. To understand why moon time is so strange—and why scientists recently created a new and unique time zone just for the moon—we have to spend a moment with Einstein.”

Image Credit: Jigar Panchal / Unsplash